scaling relations of seismic
wave energy in terms of
magnitude, distance and site
conditions
INTRODUCTION
The seismic waves generated from earthquake are body waves ( P&S)
and surface waves (love and Rayleigh).
The energy carried by these waves is seismic wave energy.
Maximum damage causing waves is S wave i.e. shear wave because
they have greater amplitude and produce vertical and horizontal
motion of the ground surface.
The classical method of response spectrum superposition is a
practical tool for design of structures expected to vibrate
without any damage during the largest possible levels of
shaking.
Response spectrum considers the SDOF.
PGA, PGV, PGD are used to explain the response of structure.
Earthquake energy is better alternative to this as it can be
calculated at the particular considered site.
OBJECTIVE
To calculate the attenuation of energy from source to
the site and the energy reaching at the considered site.
To derive empirical relations of seismic energy with
magnitude, hypocentral distance and site condition.
Wave energy at site
The energy generated at source is much larger.
But as it travels to different locations in all direction it gets attenuated.
The damage potential of waves at a location depends on the energy
reaching at site.
wave attenuation
Due to conversion of part of elastic energy of wave to heat energy due
to friction while propagation
Geometrical spreading - As the wave front moves out from the source,
the initial energy released in the earthquake is spread over an ever-
increasing area and thus the intensity of the wave decreases with
distance.
Body waves and surface waves attenuate proportional to 1/r and
1/ 𝑟 respectively.
Energy balance equation
Energy balance equation for a SDOF system can be given as:
Ekr + Ed + Ea = Eir
𝑡 𝑡 𝑡 𝑡
න 𝑚 ü ů 𝑑𝑡 + න 𝑐 ů2 𝑑𝑡 + න 𝑓 ů 𝑑𝑡 = − න 𝑚 ügů 𝑑𝑡
0 0 0 0
Input energy can be calculated based on relative motion as
internal forces within structure are computed using relative
displacement
SEISMIC WAVE ENERGY
Energy density contained in seismic wave energy is expressed as sum
of kinetic and potential wave energy.
E = Ekin + Epot
The ground motion particle displacement is
U(x,t) = A cos(ωt-kx);
1
so kinetic energy, Ekin = * ρA2 ω2 𝑠𝑖𝑛2 ωt − kx
2
1 𝜕𝑈 2
strain energy, Es = ∗ G ( )
2 𝜕𝑥
1
Potential Energy = * ρ A2 ω2 𝑠𝑖𝑛2 ωt − kx ;
2
Total energy (ET)=Kinetic Energy (Ekin)+Potential Energy (Epot)
ET = ρ A2 ω2 𝑠𝑖𝑛2 ωt − kx ;
= ρ V 2 (x,t) ( per unit area)
Energy passing per unit area of wave front = ρβV 2
Where β is the wave velocity
𝑇
Total energy (Es) = 𝜌𝛽 0 𝑣 2 𝑥, 𝑡 𝑑𝑡
DATA BASE
The data is taken from pesmos iit Roorkee.
Total 237 earthquakes data are taken and all the 3
components are considered.
Hypocentral distance is limited to 350 kms.
CORRECTION OF DATA BASE
Base line correction is applied to data by averaging out
the data so that all the data oscillates around the y=0
axis.
Base line correction is done because small drift in
acceleration from Zero axis causes a large drift of
velocity and displacement graph after integration.
STEPS OF BASELINE CORRECTION
Take the acceleration time series a(t). Distortion in
baseline has a form of quadratic function.
Integration of acceleration time series to velocity.
Fitting the quadratic function to the velocity.
V(t) = a+bt+ct2
Then compute the derivative of fitting function v(t)
𝑑𝑉(𝑡)
a`(t) =
𝑑(𝑡)
Contd.
Remove the derivative of the quadratic from the
original acceleration time history
acorrected = a(t) – a`(t)
Apply a causal, low-cut ormbsy filter with a
subjectively chosen corner frequency
Integrate to velocity and displacement.
REGRESSION ANALYSIS
Random effect regression model is used as the
heterogeneity in the site is considered.
It considers the inter event and intra event variability.
Regression model used is as :
log10(E)= C1+ C2*M- C3*log10(distance)+ C4* S
RESULTS
After regression the values of these constants C1, C2,
C3, C4 are calculated within the standard limits.
C1 = -3.622543 (0.4287088)
C2 = 1.039407 (0.1042634)
C3 = 1.363929 (0.1270322)
C4 = -0.127988 (0.0345827)
We have the formula of calculating energy at the
source as
log(E) =2*M+8.8
From the regression model to calculate the energy at
source location the distance term is taken as Zero.
By taking particular moment we can calculate the
energy by taking antilog of rest of the terms. This gives
the energy per unit area.
Contd.
The area is taken as
Area = (fault length) * (fault width)
Here,
fault length = 10^(-2.44+0.59*M) (km)
fault width = 10^(-1.01+0.32*M) (Km)
These fault length and width are taken from coppersmith (1994)
relation.
Then multiplying the energy and area values we get the total
estimated energy values. The comparison of these values is done
with the actual values.
Constant to be multiplied by energy is 2*ρ*β
ρ = 2.8
β= 3.6 * 10^15
ENERGY CALCULATION
Scaling relations
• From scaling relations of earthquake energy, magnitude and
distance for any site, by knowing magnitude only energy
reaching at a particular distance can be calculated.
• As energy produced at source is not same as of site at far
distance. So site energy is important criteria.
• By knowing energy value at site a better analysis of hazard
potential of earthquake can be done.
• By knowing energy value at site it can be used for better and safe
design of buildings and other structures.
Conclusion
The hazard assessment and quantification in terms of spectral
acceleration along with input energy is more reasonable and
useful, than using only spectral acceleration. Which is not used
very commonly.
There for developing these scaling relations of site energy is
very important.
There is availability of input energy models from various
researchers, with varying degree of validity. Hence there is a
need of validating these relations for particularly chosen sites.
references
Gutenberg B. and Richter C.F., “Earthquake Magnitude, Intensity,
Energy and Accelaration, ”
Hao Tzong-Ying,(2002)” Energy of Earthquake Response – Recent
developments” , ISET Journal of Earthquake Technology, Paper no.
416, vol. 39, No. 1-2.
Bormann P., Engdahl B. and Kind R., “Seismic Wave Propagation
and Earth models”
Chaudhary D., “Geotechnical Earthquake Engineering”, nptel ,
Department of Civil Engineering, IIT MUMBAI “
Anbazhagan P., “nptel, introduction to engineering seismology “
Khashaee P., Mohraz B., Sadek F., Lew H.S., Gross J L.,
“Distribution of Earthquake Input Energy in Structures”, United
States Department of Commerce Technology Administration,
National Institute of Standards and Technology, NISTIR 6903
Have a wonderful day
